As portable electric products such as a notebook, a video camera, a mobile phone, and the like, have been actively used, the importance of a rechargeable battery mainly used as a driving power supply therefor is being increased. In recent years, due to the increased focus on environmental problems, a research for a hybrid vehicle and an electric vehicle is being conducted actively. Since a hybrid vehicle or an electric vehicle drives vehicles by using charging and discharging energy of a battery, the hybrid or electric vehicle may have more excellent fuel efficiency and reduce pollutants, as compared with a vehicle using only an engine, and therefore has received a positive consumer response. Subsequently, an interest and a research for a battery, which is a core part of the hybrid vehicle or the electric vehicle, has been intensively increased.
In addition, the importance for a technology of a battery management system for more efficiently using and managing a battery has increased. In particular, the battery management system needs to accurately predict a state-of-health (SOH) of a battery so as to appropriately control a use strategy of a charging or discharging output and a state of charging (SOC) of the battery.
Therefore, in order to increase prediction accuracy of the SOH of a battery to reduce an estimated error, there is a need to obtain an average value closest to an actual SOH except for values that are predicted higher or lower than the actual SOH.